Proof by Contradiction vs Proof by Contrapositive — Not Interchangeable

The confusion: "My teacher says proof by contradiction and proof by contrapositive are the same thing. I see them written differently but used interchangeably. Are they? What is going on?"

They are not the same, although they are close cousins. For a conditional statement "if P then Q," there is a genuine structural difference in what each proof assumes, what it derives, and what tools it is allowed to use along the way. The proofs often look similar because they end up at the same conclusion, but their middles are different. A teacher using the terms as synonyms is being loose; the textbook distinction is real.

This article lays down the difference precisely and shows where it bites.

The two techniques, stated exactly

You want to prove the conditional statement P \Rightarrow Q ("if P, then Q").

Proof by contrapositive. Prove instead the contrapositive: \lnot Q \Rightarrow \lnot P. That is, assume \lnot Q and derive \lnot P. You never assume P at all. The contrapositive is logically equivalent to the original (see Converse vs Contrapositive), so proving it proves the original.

Proof by contradiction. Assume the negation of the whole implication — that is, assume P \land \lnot Q ("P is true and Q is false"). Derive any contradiction whatsoever. The contradiction rules out P \land \lnot Q, so P \Rightarrow Q must hold.

Compare what each one assumes:

Contrapositive assumes just \lnot Q. Derives \lnot P. The derivation has to land exactly at \lnot P — nowhere else.
Contradiction assumes both P and \lnot Q. Derives any contradiction. The derivation can collide with anything — a prior theorem, a definition, or the assumption itself.

Contradiction gives you more premises to work with (you have P in hand, not just \lnot Q) and more targets to hit (any contradiction counts, not specifically \lnot P). Contrapositive gives you a tighter, more directed proof with fewer tools.

A worked comparison

Claim: If n^2 is even, then n is even. (Here P = "n^2 is even", Q = "n is even".)

By contrapositive. Prove \lnot Q \Rightarrow \lnot P: if n is odd, then n^2 is odd.

Let n be odd. Then n = 2k + 1 for some integer k. So n^2 = (2k+1)^2 = 4k^2 + 4k + 1 = 2(2k^2 + 2k) + 1, which is odd. Done.

Notice: we never assumed n^2 is even. We only assumed n is odd. The derivation proceeded forward to n^2 being odd.

By contradiction. Assume P \land \lnot Q: suppose n^2 is even and n is odd. Then n = 2k + 1 and n^2 = 4k^2 + 4k + 1, which is odd — contradicting that n^2 is even.

Notice: we used both parts of the assumption. The assumption "n^2 is even" plus the derived "n^2 is odd" is the collision.

Are they the same proof? Functionally, very nearly. The algebra is identical. But the contrapositive proof has a cleaner logical form: it is a direct forward argument from "n is odd" to "n^2 is odd," with no assumption about n^2 ever entering. The contradiction proof uses the extra assumption "n^2 is even" explicitly at the very last step.

When contradiction does strictly more than contrapositive

There are proofs that can be done by contradiction but cannot be easily done by contrapositive. These are proofs where the contradiction that arises is not of the form "P is false" but of some other form.

Example. Prove: if x and y are positive reals, then x + y \geq 2\sqrt{xy}.

By contradiction. Suppose x, y > 0 and x + y < 2\sqrt{xy}. Squaring both sides (allowed because both are positive): (x + y)^2 < 4xy, i.e. x^2 + 2xy + y^2 < 4xy, so x^2 - 2xy + y^2 < 0, i.e. (x - y)^2 < 0. But a square of a real number is never negative — contradiction.

The contradiction here is "a square is negative," which is not the negation of "x and y are positive reals." The proof uses the full combination (x, y positive AND the inequality fails) to derive a collision with a general fact about real numbers ((x - y)^2 \geq 0). A contrapositive proof would have to go the other way: assume (x - y)^2 \geq 0 (or equivalently x + y \geq 2\sqrt{xy}) and derive x, y > 0 — but that is circular and not what we want.

Lesson: when the natural contradiction is not the negation of the hypothesis but some general mathematical fact, proof by contradiction is the cleaner tool. Contrapositive would force you to contort the argument.

When contrapositive is preferable

Example. Prove: if n^3 + 5 is odd, then n is even. (P = "n^3 + 5 odd", Q = "n even".)

By contrapositive. Assume n is odd. Then n^3 is odd (product of three odds), and n^3 + 5 is odd + odd = even. Done — that is \lnot P.

By contradiction. Assume n^3 + 5 is odd and n is odd. Then n^3 is odd, n^3 + 5 is odd + odd = even — contradicting n^3 + 5 odd.

Same algebra. But the contrapositive version is cleaner because it never needed the premise "n^3 + 5 is odd" — it derived \lnot P directly from \lnot Q and let the logical equivalence do the work. The contradiction version drags in the premise "n^3 + 5 is odd" and uses it at the end as the collision partner. This is clutter.

Why the distinction sharpens your writing: contrapositive proofs tend to be tighter because you are using the minimum information. Contradiction proofs tend to be more flexible because you are allowed to assume more. For an implication P \Rightarrow Q where the clean path is "if not-Q, here is a forward derivation of not-P," contrapositive reads better. For one where the contradiction will come from some external fact, contradiction is the natural fit.

What your teacher probably means

When a teacher says "they are the same," they usually mean one of two things:

At the level of which theorems get proved, yes — proving P \Rightarrow Q by contrapositive and by contradiction both produce P \Rightarrow Q as a theorem. Both techniques are logically valid in classical mathematics, and both give the same final conclusion.
A contrapositive proof can always be rewritten as a contradiction proof by adding "suppose P" at the top and using the contradiction between derived \lnot P and assumed P. This is true. Every contrapositive proof is essentially a special case of contradiction where the collision is "P is true and \lnot P is derived."

But the converse is not true — not every contradiction proof can be rewritten as a contrapositive. If the collision is with a general mathematical fact (like (x - y)^2 \geq 0 above), you cannot shoehorn that into contrapositive form without adding extra machinery.

So the two techniques are not fully interchangeable; they are in a one-way relationship. Contrapositive is a stricter discipline inside the broader practice of contradiction.

Classify the proof style

Claim: If n^2 + n is odd, then no solution exists.

Wait — this is not a well-formed conditional. Let's try: If n is a positive integer, then n^2 + n is even.

Here P = "n is a positive integer", Q = "n^2 + n is even."

Attempt 1 (contrapositive). Assume \lnot Q: n^2 + n is odd. Then... try to derive \lnot P: "n is not a positive integer." This is hard — what goes wrong? The trouble is that n^2 + n = n(n+1) is always even for any integer n, positive or not. So n^2 + n being odd is already a contradiction with the mathematics, not a clean path to \lnot P.

Attempt 2 (contradiction). Assume P (n is a positive integer) and \lnot Q (n^2 + n is odd). Then n(n+1) is odd. But n and n+1 are consecutive integers, so one of them is even, and the product of an even with anything is even — contradicting "n(n+1) is odd."

Result: contradiction works cleanly. Contrapositive is awkward because the "wrong" state of Q is already impossible in isolation — you do not need \lnot P to reach trouble, and contrapositive requires you to derive \lnot P specifically.

Takeaway: when the hypothesis is doing real work — when \lnot Q alone would be impossible, so you need P to get traction — contradiction is the right tool. Contrapositive wants \lnot Q to be a consistent starting point from which \lnot P follows.

What to remember

Contrapositive: assume \lnot Q, derive \lnot P. Forward proof with tighter premises.
Contradiction: assume P \land \lnot Q, derive any contradiction. Broader premises, broader targets.
Every contrapositive proof is a contradiction proof in disguise; not every contradiction proof can be recast as contrapositive.
In informal use, the terms get blurred. In careful writing — and in JEE free-response problems where the grader cares about the structure — state which technique you are using, and use the more appropriate one for the situation.